Examinations

How accurate must a boundary survey be, and what does meeting the state standard prove?

A boundary survey has to meet the accuracy standard of the state where the land lies, and the numbers differ by state. Washington's tolerance between two positions is 0.07 feet plus 200 parts per million. Virginia requires a traverse to close within 1 part in 10,000 in rural areas and 1 in 20,000 in urban areas. Arkansas sets 0.25 feet between urban corners. Meeting the standard proves the measurements, not the location of the line.

Gloved hands press a closed brass survey case against a wooden stake, a field notebook on the grass

What does an accuracy standard for a boundary survey measure?

How closely the surveyed positions agree with each other or with the control the survey is tied to, at a stated confidence level, and how well a measured traverse closes on itself.

Washington's rule, WAC 332-130-085, uses relative accuracy. It measures how closely two positions shown on a boundary survey agree with each other, using the distance between them as drawn on the map. The check is not about where either point sits on the earth, only about the two of them together. The tolerance is stated at "the ninety-five percent confidence level."

Missouri's rule, 20 CSR 2030-16.040, uses two measures. One is the random-error uncertainty of any direction or distance shown on the plat. The other is the positional uncertainty of the plat coordinates relative to the control the surveyor held fixed. Both limits apply at "the sixty-eight percent (68%) confidence level (one (1) sigma)." The rule also requires the surveyor to "make an effort to detect and remove systematic errors."

The confidence level changes what a number means. A tolerance stated at one sigma is met by 68 percent of expected results, and one stated at 95 percent by 95 percent of them. A Missouri limit and a Washington limit with the same figure in feet are therefore not the same requirement.

The Bureau of Land Management, in its October 2019 standards for cadastral surveys with satellite positioning, uses two measures: local accuracy, relative to nearby connected points, and network accuracy, relative to the National Spatial Reference System. Each is the semi-major axis of the 95 percent error ellipse.

Which accuracy standard applies to a particular survey?

The minimum standard of the state where the land is, for the class of property the state assigns. Federal cadastral surveys follow BLM standards. Where a local requirement is stricter, Arkansas applies the more stringent one.

Each of the four states cited here sets its standard by administrative rule: Washington and Virginia in their administrative codes, Missouri in its Code of State Regulations and Arkansas in Standards of Practice No. 1. Washington describes its tolerance as "minimum standards only" and says it may be applied to boundary surveys that use field traverses.

Classes matter. Virginia, in 18VAC10-20-370, sets one group of figures for rural areas and another for urban and suburban areas. Arkansas Standards of Practice No. 1, in force since August 7, 2020, has four classes: urban (Class A), suburban (Class B), rural (Class C) and mountain or marsh (Class D). Section 3.1-C adds that where local jurisdictional requirements conflict with the state standards, "the more stringent shall apply." Under Section 5.1 the state board that licenses engineers and surveyors enforces them.

Washington's rule says higher accuracy is expected where property values are higher or other circumstances call for it, without giving a separate figure. A survey of a commercial lot in Seattle and one of timberland can meet the same written standard while the surveyor is expected to do more on the commercial lot.

The plat or report should name the standard it claims to meet. A survey that names none leaves the reviewing surveyor to work out which class applied and to test the work against that class.

How do the state accuracy numbers compare?

Washington: 0.07 feet plus 200 ppm at 95 percent. Virginia: 0.05 feet plus 30 ppm in urban areas. Missouri: 50 ppm or 0.10 foot at one sigma in urban areas. Arkansas: 0.25 feet between urban corners.

Washington: 0.07 feet plus 200 parts per million between any two positions, at the 95 percent confidence level, with no separate classes. On a line 300 feet long the 200 ppm part adds 0.06 feet, so the two corners may disagree by 0.13 feet.

Virginia, rural areas: neither axis of the error ellipse for any control point, property corner or independent point may exceed 0.10 foot plus 50 ppm at the 95 percent confidence level. Urban and suburban areas: 0.05 foot plus 30 ppm. On the same 300-foot line the urban figure is about 0.059 feet.

Missouri: for urban property, the uncertainty may not exceed 50 ppm or 0.10 foot for distances under 2,000 feet. For rural property, 100 ppm or 0.10 foot for distances under 1,000 feet. Both at one sigma.

Arkansas: the maximum positional tolerance of property corners with respect to each other is plus or minus 0.25 feet for urban Class A, 0.50 feet for suburban Class B, 0.75 feet for rural Class C and 1.50 feet for mountain or marsh Class D. The rule measures it as the horizontal distance between the measured position of a monument and its computed location.

The Utah Council of Land Surveyors adopted a model standard on August 14, 1999 with four classes of positional tolerance: urban 0.07 feet (20 mm) plus 50 ppm, suburban 0.13 feet (40 mm) plus 100 ppm, rural 0.26 feet (80 mm) plus 200 ppm, and mountain or marshland 0.66 feet (200 mm) plus 200 ppm. It is a professional model, not a state rule.

What is error of closure, and what ratios do the standards require?

When a traverse is run around a parcel and back to its starting point, the error of closure is the gap between the computed end and the start. Virginia's limits are 1 in 10,000 rural and 1 in 20,000 urban.

Virginia states the closure limits directly. In rural areas the maximum traverse error of closure is one part in 10,000, and angular closure may not exceed 20 seconds times the square root of the number of angles turned. In urban and suburban areas the limit is one foot per 20,000 feet of perimeter, and angular closure may not exceed 10 seconds times the square root of the number of angles. Instruments must read directly to at least six seconds of arc.

A rural Virginia parcel with a perimeter of 1,200 feet may therefore close with a linear error of no more than 0.12 feet. The same parcel in a suburban area may close with no more than 0.06 feet.

The 1999 Utah model sets closure by class as well: 1:15,000 urban, 1:10,000 suburban, 1:7,500 rural and 1:5,000 for mountain or marshland, with angular closure of 10 seconds times the square root of the number of stations in the urban class. Its Table 2 adds that "All requirements of each class must be satisfied in order to qualify for that particular class of survey."

A closure ratio says the measurements are consistent with one another. It does not say whether the traverse began on the right monument. A survey can close at 1 in 20,000 around the wrong corner.

Can a GPS survey meet the standard with a single measurement?

Not in Virginia. Its rule calls a single RTK vector a no-check solution that does not meet the minimum standards by itself. BLM standards for federal cadastral corners require local accuracy under 0.050 meters at 95 percent.

Under 18VAC10-20-370(E), multiple simultaneous observations are required to meet the closure and accuracy standards when satellite positioning is used. A single real-time kinematic vector solution is a "no check" solution, does not meet minimum standards by itself and may not be used without additional independent checks. Virginia also caps the positional uncertainty of independent boundary corners set with GPS at 0.07 foot, or 20 mm plus or minus 50 ppm, at the 95 percent confidence level.

The BLM standards for cadastral surveys with satellite positioning (GNSS), developed with the USDA Forest Service and dated October 2019, give four figures. Local accuracy: less than 0.025 meters for project control and less than 0.050 meters for corner measurements. Network accuracy: less than 0.050 meters for project control and less than 0.100 meters for corner measurements.

In a dispute, the question for a GNSS survey is whether the field files show more than one independent occupation of each disputed corner. If the opposing surveyor located a corner once with RTK and checked nothing, the Virginia rule treats that position as unchecked.

What must the surveyor do if the accuracy standard is not met?

In Washington the surveyor states the reasons, identifies each monument whose position exceeds the standard and gives the amount. Virginia requires the plat to say whether it rests on a current field survey or a compilation.

WAC 332-130-085 accepts that the tolerance cannot always be met. Relief, vegetation, improvements and the size or shape of the property are listed as possible causes. When the tolerance is exceeded, the surveyor gives the reasons, identifies each monument whose position exceeds the standard and states by how much.

Virginia's rule addresses what the plat says about its own basis. Under 18VAC10-20-370(F)(2)(r) the plat states whether the survey is based on a current field survey, including closed-loop traverse methods such as total station or redundant GNSS observations, or on a compilation from deeds, plats, surveys by others or a combination. A compilation must show in its title that it does not represent a current land boundary survey, and it may not be titled as one.

The same rule requires bearings to the nearest second of arc, distances to the nearest 0.01 foot, and area to the nearest 0.01 acre for rural property or the nearest square foot for urban and suburban property. Inconsistencies found in researching common boundaries with adjoining land are to be clearly noted on the plat.

A plat that is silent on its basis, on exceeded tolerances or on inconsistencies with the adjoiners gives the reviewer the first set of questions for the surveyor's deposition.

Does meeting the accuracy standard prove that the line is in the right place?

No. Accuracy standards test measurements. Where the line runs depends on the evidence the surveyor followed, and in Morrow v. White a survey that could vary 15 feet in any direction did not prove the boundary.

Morrow v. White, 12 Ark. App. 16, 670 S.W.2d 459, was decided by the Arkansas Court of Appeals on June 13, 1984 (No. CA 83-287). The adjoining lands came from the 1848 General Land Office survey, and the common boundary was the original government line. The chancellor found that the appellant had not proved where it ran, and the court affirmed.

The court compared the relocation with the original field notes. One segment was run 1,314.46 feet where the field notes called for 1,320 feet, and the resulting southwest corner of Section 18 sat about 20 feet north of the original point. The surveyor had not tied his work to a known corner to the east, estimating that it would cost more than $10,000, and the chancellor found that the survey could vary "as much as 15 feet in any direction." The court held that this did not meet the certainty required for a boundary between adjoining owners.

The opinion restates the rule for retracement: later surveys retrace the original surveyor's steps from the field notes and plats and do not correct his errors. It also held that yielding to stone piles was error where the piles were not shown to sit where the plat and field notes placed them. The surveyor acknowledged the 1973 BLM Manual of Surveying Instructions as accepted professional practice for relocating lost corners.

Two surveys that each meet a state tolerance can still put the line in different places if they started from different monuments or read the deed differently. The accuracy check comes first because it separates measurement noise from a disagreement about evidence.

What does a review of the opposing survey check about accuracy?

Which standard and class the survey claims, the confidence level, closure and angular figures, redundancy of GNSS observations, the control held fixed, and whether exceeded tolerances and conflicts with adjoiners are disclosed.

The reviewer starts with the plat: the standard named, the class of property, the basis of the survey and the date of the fieldwork. If no standard is named, the reviewer identifies the state rule and the class that applies to the parcel.

Field data come next. For a traverse, the reviewer asks for the raw angle and distance observations and the closure computation. For a satellite survey, the reviewer asks for the vector files and the number of independent observations of each corner. Virginia's no-check rule and the BLM figures give the test.

The reviewer then computes the tolerance for each disputed line. If the two surveys differ at a corner by less than the combined tolerance of both, the difference is within the measurement precision. If they differ by feet, as in Morrow, the cause lies in the evidence used, and the review turns to monuments, field notes and deeds.

Missouri's requirement to detect and remove systematic errors adds one more question: whether the surveyor checked the instruments, the control and any scale factor before computing the corners.

What does IAE prepare when survey accuracy is disputed?

An expert report on the client's questions about deed lines, overlaps and encroachments, or a review of the opposing surveyor's report, issued within up to 5 business days after the format and cost are confirmed.

The boundary survey page covers boundary, quiet title and encroachment cases in US courts: where the deed line runs on the ground, whether two descriptions overlap, whether a fence or a building encroaches. The surveyor answers the questions agreed with the client, working from the record documents and a field survey. The client sends the deeds, plats and any prior surveys with the request, and for an accuracy dispute the field data and closure computations of each survey if the client has them.

The secretariat confirms receipt the same day. Within one business day IAE confirms the format, the cost, the deadline and the specialist from its registry, and the client approves the candidate and the CV before work starts. The signed PDF is issued in up to 5 business days, and cases with tight deadlines have a 48-hour rush.

Every specialist is a member of the public IAE registry and passed the council's attestation under the published Criteria Matrix. The Member ID in the document opens the specialist's category, specialization, attestation date and record status on the status check page or by QR code. The specialist confirms in the document that he or she has no interest in the outcome of the case.

Materials go through the contact form and are not passed to third parties without the client's consent. Neither IAE nor the surveyor guarantees the outcome or influences the decision. IAE is not a law firm and does not advise on case strategy.

Sources

Read also

All articles

Expert examinations

Ready to discuss your case?

Send the materials: the secretariat will confirm the format, cost, deadline and the specialist within one business day.

Expertise request

Received

Cost and timeline estimate

WorkReview of an opinion
TimelineUp to 5 business days
ExpertIAE registry member